Skip to content

Benchmark

Long Horizon Terminal Bench

Scores are percentages. Higher is better.

Models measured
9
Scores recorded
9
Sources
1
IQ weight
0.56

Score distribution

2.3%
Mean
2%
Median
1.6%
Std dev
1%
10th percentile
3%
90th percentile
0%
Min
0%6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.56 for chat models, measured over 9 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Kimi K3· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim6%
2DeepSeek V4 Pro· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim3%
3Kimi K2.7 Code· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim3%
4MiniMax M3· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim3%
5DeepSeek V4 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim2%
6GLM 5.1· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim2%
7GLM 5.2· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim1%
8Hy3 (Deep Infra)· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim1%
9Kimi K2.6· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim0%

Data credits: Hugging Face official benchmark leaderboards; licence varies by benchmark